Pub Date : 2014-12-08DOI: 10.1109/ICNC.2014.6975961
Wei Zhong, Li Fang, Long Ye, Qin Zhang, Siqi Shi
In this paper, we extend the partial modulation technique to obtain the highly desired linear-phase (LP) characteristics of oversampled nonuniform filter banks (NUFBs), which makes it possible to design LP oversampled NUFBs based on the efficient modulation technique. Further for the subbands with sampling factors violating the guard band constraint, a phase modification structure is also proposed to avoid uneliminable large aliasing and meanwhile maintain the LP characteristics of analysis/synthesis filters, realizing arbitrary integer decimation. By using the proposed algorithm, the constraints on the LP oversampled NUFB design are simplified into those only imposed on several prototype filters, largely reducing the design complexity. As demonstrated by examples, the proposed algorithm can achieve LP oversampled NUFBs with arbitrary integer decimation in a simple and efficient way.
{"title":"Design of linear-phase oversampled nonuniform filter banks with arbitrary integer sampling factors","authors":"Wei Zhong, Li Fang, Long Ye, Qin Zhang, Siqi Shi","doi":"10.1109/ICNC.2014.6975961","DOIUrl":"https://doi.org/10.1109/ICNC.2014.6975961","url":null,"abstract":"In this paper, we extend the partial modulation technique to obtain the highly desired linear-phase (LP) characteristics of oversampled nonuniform filter banks (NUFBs), which makes it possible to design LP oversampled NUFBs based on the efficient modulation technique. Further for the subbands with sampling factors violating the guard band constraint, a phase modification structure is also proposed to avoid uneliminable large aliasing and meanwhile maintain the LP characteristics of analysis/synthesis filters, realizing arbitrary integer decimation. By using the proposed algorithm, the constraints on the LP oversampled NUFB design are simplified into those only imposed on several prototype filters, largely reducing the design complexity. As demonstrated by examples, the proposed algorithm can achieve LP oversampled NUFBs with arbitrary integer decimation in a simple and efficient way.","PeriodicalId":208779,"journal":{"name":"2014 10th International Conference on Natural Computation (ICNC)","volume":"40 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128022633","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-12-08DOI: 10.1109/ICNC.2014.6975908
Jinbao Zhang, Ming Liu, Yongqiang Zhao, Xingguo Lu
The probabilistic characteristics of components can't be completely expressed by the S-N curve with parameters estimated by small number of specimens. Particle Swarm Optimization (PSO) is introduced to fit parameters with the incomplete test data, which can take advantage of the entire information of the specimens to obtain the globally optimal solution. With the fitness function offered based on the principle of the total minimum mean-square value of fitting errors, the parameters of the three-parameter P-S-N curve are estimated with PSO. In sequence, the obtained P-S-N curve is applied in the fatigue damage accumulation model for reliability prediction. The above models are verified with test data with relation to two different 45 steels. The simulation results match well with experiment data.
{"title":"P-S-N curves with parameters estimated by particle swarm optimization and reliability prediction","authors":"Jinbao Zhang, Ming Liu, Yongqiang Zhao, Xingguo Lu","doi":"10.1109/ICNC.2014.6975908","DOIUrl":"https://doi.org/10.1109/ICNC.2014.6975908","url":null,"abstract":"The probabilistic characteristics of components can't be completely expressed by the S-N curve with parameters estimated by small number of specimens. Particle Swarm Optimization (PSO) is introduced to fit parameters with the incomplete test data, which can take advantage of the entire information of the specimens to obtain the globally optimal solution. With the fitness function offered based on the principle of the total minimum mean-square value of fitting errors, the parameters of the three-parameter P-S-N curve are estimated with PSO. In sequence, the obtained P-S-N curve is applied in the fatigue damage accumulation model for reliability prediction. The above models are verified with test data with relation to two different 45 steels. The simulation results match well with experiment data.","PeriodicalId":208779,"journal":{"name":"2014 10th International Conference on Natural Computation (ICNC)","volume":"286 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134293765","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-12-08DOI: 10.1109/ICNC.2014.6975806
Cong Zhan, Shihua Chen
This paper further discusses the global asymptotic stability of Cohen-Grossberg neural networks with time delays. A approach to study the global asymptotic stability of neural networks is proposed. Better test conditions are obtained and a numerical simulation is given to demonstrate the effectiveness of the criterion.
{"title":"Global asymptotic stability analysis of Cohen-Grossberg neural networks with delays","authors":"Cong Zhan, Shihua Chen","doi":"10.1109/ICNC.2014.6975806","DOIUrl":"https://doi.org/10.1109/ICNC.2014.6975806","url":null,"abstract":"This paper further discusses the global asymptotic stability of Cohen-Grossberg neural networks with time delays. A approach to study the global asymptotic stability of neural networks is proposed. Better test conditions are obtained and a numerical simulation is given to demonstrate the effectiveness of the criterion.","PeriodicalId":208779,"journal":{"name":"2014 10th International Conference on Natural Computation (ICNC)","volume":"93 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134395431","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-12-08DOI: 10.1109/ICNC.2014.6975839
Yu-an Zhang, Qinglian Ma, Hiroshi Furutani
Markov chain is a powerful tool for analyzing the evolutionary process of a stochastic system. To select GA parameters such as mutation rate and population size are important in practical application. The value of this parameter has a big effect on the viewpoint of Markov chain. In this paper, we consider properties of stationary distribution with mutation in GAs. We used Markov chain to calculate distribution. If the population is in linkage equilibrium, we used Wright-Fisher model to get the distribution of first order schema. We define the mixing time is the time to arrive stationary distribution. We adopt Hunter's mixing time to estimate the mixing time Tm of the first order schema.
{"title":"Markov chain model of schema evolution and its application to stationary distribution","authors":"Yu-an Zhang, Qinglian Ma, Hiroshi Furutani","doi":"10.1109/ICNC.2014.6975839","DOIUrl":"https://doi.org/10.1109/ICNC.2014.6975839","url":null,"abstract":"Markov chain is a powerful tool for analyzing the evolutionary process of a stochastic system. To select GA parameters such as mutation rate and population size are important in practical application. The value of this parameter has a big effect on the viewpoint of Markov chain. In this paper, we consider properties of stationary distribution with mutation in GAs. We used Markov chain to calculate distribution. If the population is in linkage equilibrium, we used Wright-Fisher model to get the distribution of first order schema. We define the mixing time is the time to arrive stationary distribution. We adopt Hunter's mixing time to estimate the mixing time Tm of the first order schema.","PeriodicalId":208779,"journal":{"name":"2014 10th International Conference on Natural Computation (ICNC)","volume":"70 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130685354","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-12-08DOI: 10.1109/ICNC.2014.6975823
Qunjiao Zhang, Juan Luo, Jie Liu
In this paper, the robust consensus between two FitzHugh-Nagumo networks with external disturbances is investigated, based on the sliding mode control method. Some synchronization criterion and theoretical ultimate error bounds are derived to realize the robust synchronization. Finally, some numerical simulations are given to illustrate the validity of the proposed results.
{"title":"Robust consensus of FitzHugh-Nagumo networks with disturbances via sliding mode control","authors":"Qunjiao Zhang, Juan Luo, Jie Liu","doi":"10.1109/ICNC.2014.6975823","DOIUrl":"https://doi.org/10.1109/ICNC.2014.6975823","url":null,"abstract":"In this paper, the robust consensus between two FitzHugh-Nagumo networks with external disturbances is investigated, based on the sliding mode control method. Some synchronization criterion and theoretical ultimate error bounds are derived to realize the robust synchronization. Finally, some numerical simulations are given to illustrate the validity of the proposed results.","PeriodicalId":208779,"journal":{"name":"2014 10th International Conference on Natural Computation (ICNC)","volume":"23 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133676967","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-12-08DOI: 10.1109/ICNC.2014.6975987
Mingxin Hou, Li Jiang, M. Jin, Hong Liu, Zhaopeng Chen
A multi-finger dynamics model has been presented in this study, which contains a single finger dynamics model equation and a restraint equation between fingers based on Lagrangian multiplier controller. To validate the model, an EtherCAT master and slave platform has been developed based on FPGA. Meanwhile, the multi-finger dynamics algorithm has been designed in the TwinCAT. Finally, the experiments demonstrate this strategy can be implemented and operated by online grasping object.
{"title":"Analysis of the multi-finger dynamics for robot hand system based on EtherCAT","authors":"Mingxin Hou, Li Jiang, M. Jin, Hong Liu, Zhaopeng Chen","doi":"10.1109/ICNC.2014.6975987","DOIUrl":"https://doi.org/10.1109/ICNC.2014.6975987","url":null,"abstract":"A multi-finger dynamics model has been presented in this study, which contains a single finger dynamics model equation and a restraint equation between fingers based on Lagrangian multiplier controller. To validate the model, an EtherCAT master and slave platform has been developed based on FPGA. Meanwhile, the multi-finger dynamics algorithm has been designed in the TwinCAT. Finally, the experiments demonstrate this strategy can be implemented and operated by online grasping object.","PeriodicalId":208779,"journal":{"name":"2014 10th International Conference on Natural Computation (ICNC)","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132099200","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-12-08DOI: 10.1109/ICNC.2014.6975936
Qian Chen, Wenhao Zhu, Chaoyou Ju, Wu Zhang
One of the key problems of information extraction is to design a cross domain extraction procedure that can adapt different domain topics and text formats. However, most information extraction methods focus on specific areas or only have limited scalability for semi-structured texts. We argue that the problem of cross domain information extraction is basically introduced by domain related features. For example, the features used for price extraction in e-commerce websites cannot be directly applied in the case of extracting salary for recruiting websites. In worst case, a whole extraction model is required to be implemented despite the fact that there are common characters for price and salary. In this paper we propose a cross domain solution by dismantling domain relevant features into sub-features that are less domain related. The sub-features include composite features (those can be represented with a combination of several other features) and atomic features (features that can't be dismantled). To manage the features effectively we propose a multi-level feature model by organizing the features as well as their relations. With this model, we give an information extraction method that can be quickly shifted when the extraction domain changes.
{"title":"Cross domain web information extraction with multi-level feature model","authors":"Qian Chen, Wenhao Zhu, Chaoyou Ju, Wu Zhang","doi":"10.1109/ICNC.2014.6975936","DOIUrl":"https://doi.org/10.1109/ICNC.2014.6975936","url":null,"abstract":"One of the key problems of information extraction is to design a cross domain extraction procedure that can adapt different domain topics and text formats. However, most information extraction methods focus on specific areas or only have limited scalability for semi-structured texts. We argue that the problem of cross domain information extraction is basically introduced by domain related features. For example, the features used for price extraction in e-commerce websites cannot be directly applied in the case of extracting salary for recruiting websites. In worst case, a whole extraction model is required to be implemented despite the fact that there are common characters for price and salary. In this paper we propose a cross domain solution by dismantling domain relevant features into sub-features that are less domain related. The sub-features include composite features (those can be represented with a combination of several other features) and atomic features (features that can't be dismantled). To manage the features effectively we propose a multi-level feature model by organizing the features as well as their relations. With this model, we give an information extraction method that can be quickly shifted when the extraction domain changes.","PeriodicalId":208779,"journal":{"name":"2014 10th International Conference on Natural Computation (ICNC)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124306477","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-12-08DOI: 10.1109/ICNC.2014.6975807
Xin Xu, Jie Li, Huiling Chen
Proper parameter settings of support vector machine (SVM) and feature selection are of great importance to its efficiency and accuracy. In this paper, we propose a parallel adaptive particle swarm optimization algorithm to simultaneously perform the parameter optimization and feature selection for SVM, termed PTVPSO-SVM. It is implemented in an efficient parallel environment using PVM (Parallel Virtual Machine). In the proposed method, a weighted function is adopted to design the objective function of PSO, which takes into account the average accuracy rates (Acc), the number of support vectors (SVs) and the selected features simultaneously. The adaptive control parameters including the time varying acceleration coefficients (TVAC) and inertia weight (TVIW) are employed to efficiently control the local and global search in PSO and mutation operators are introduced to overcome the problem of the premature convergence of PSO algorithm. The experimental results clearly confirm the superiority of the proposed method over the other two reference methods on several real world datasets. It also reveals that the PTVPSO-SVM can not only obtain much more appropriate model parameters, discriminative feature subset as well as smaller sets of SVs but also significantly reduce the computational time, giving high predictive accuracy.
{"title":"Enhanced support vector machine using parallel particle swarm optimization","authors":"Xin Xu, Jie Li, Huiling Chen","doi":"10.1109/ICNC.2014.6975807","DOIUrl":"https://doi.org/10.1109/ICNC.2014.6975807","url":null,"abstract":"Proper parameter settings of support vector machine (SVM) and feature selection are of great importance to its efficiency and accuracy. In this paper, we propose a parallel adaptive particle swarm optimization algorithm to simultaneously perform the parameter optimization and feature selection for SVM, termed PTVPSO-SVM. It is implemented in an efficient parallel environment using PVM (Parallel Virtual Machine). In the proposed method, a weighted function is adopted to design the objective function of PSO, which takes into account the average accuracy rates (Acc), the number of support vectors (SVs) and the selected features simultaneously. The adaptive control parameters including the time varying acceleration coefficients (TVAC) and inertia weight (TVIW) are employed to efficiently control the local and global search in PSO and mutation operators are introduced to overcome the problem of the premature convergence of PSO algorithm. The experimental results clearly confirm the superiority of the proposed method over the other two reference methods on several real world datasets. It also reveals that the PTVPSO-SVM can not only obtain much more appropriate model parameters, discriminative feature subset as well as smaller sets of SVs but also significantly reduce the computational time, giving high predictive accuracy.","PeriodicalId":208779,"journal":{"name":"2014 10th International Conference on Natural Computation (ICNC)","volume":"468 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124384389","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Particle swarm optimization (PSO) algorithm has attracted great attention as a stochastic optimizing method due to its simplicity and power strength in optimization fields. However, two issues are still to be improved, especially, for complex multimodal problems. One is the premature convergence for multimodal problems. The other is the low efficiency for complex problems. To address these two issues, firstly, a strategy based on the global optimum prediction is proposed. A predicting model is established on the low-dimensional feature space with the principle component analysis technique, which has the ability to predict the global optimal position by the feature reflecting the evolution tendency of the current swarm. Then the predicted position is used as a guideline exemplar of the evolution process together with pbest and gbest. Secondly, a strategy, called adaptive mutation, is proposed, which can evaluate the crowding level of the aggregating particle swarm by using the distribution topology of each dimension, and hence, can get the possible location of local optimums and escape from the valleys with the generalized non-uniform mutation operator subsequently. The performance of the proposed global prediction-based adaptive mutation particle swarm optimization (GPAM-PSO) is tested on 8 well-known benchmark problems, compared with 9 existing PSO in terms of both accuracy and efficiency. The experimental results demonstrate that GPAM-PSO outperforms all reference PSO algorithms on both the solution quality and convergence speed.
{"title":"Global prediction-based adaptive mutation particle swarm optimization","authors":"Qiuying Li, Gaoyang Li, Xiaosong Han, Jianping Zhang, Yanchun Liang, Binghong Wang, Hong Li, Jinyu Yang, Chunguo Wu","doi":"10.1109/ICNC.2014.6975846","DOIUrl":"https://doi.org/10.1109/ICNC.2014.6975846","url":null,"abstract":"Particle swarm optimization (PSO) algorithm has attracted great attention as a stochastic optimizing method due to its simplicity and power strength in optimization fields. However, two issues are still to be improved, especially, for complex multimodal problems. One is the premature convergence for multimodal problems. The other is the low efficiency for complex problems. To address these two issues, firstly, a strategy based on the global optimum prediction is proposed. A predicting model is established on the low-dimensional feature space with the principle component analysis technique, which has the ability to predict the global optimal position by the feature reflecting the evolution tendency of the current swarm. Then the predicted position is used as a guideline exemplar of the evolution process together with pbest and gbest. Secondly, a strategy, called adaptive mutation, is proposed, which can evaluate the crowding level of the aggregating particle swarm by using the distribution topology of each dimension, and hence, can get the possible location of local optimums and escape from the valleys with the generalized non-uniform mutation operator subsequently. The performance of the proposed global prediction-based adaptive mutation particle swarm optimization (GPAM-PSO) is tested on 8 well-known benchmark problems, compared with 9 existing PSO in terms of both accuracy and efficiency. The experimental results demonstrate that GPAM-PSO outperforms all reference PSO algorithms on both the solution quality and convergence speed.","PeriodicalId":208779,"journal":{"name":"2014 10th International Conference on Natural Computation (ICNC)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114399021","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2014-12-08DOI: 10.1109/ICNC.2014.6975813
Mei Liu, Ran Yang, Bolin Liao
The online solution of optimization (including minimization and maximization) is viewed as a basic and important issue, which has been widely arisen in scientific researches and engineering applications. In this paper, a new recurrent neural network (NRNN) is generalized and investigated for the nonlinear optimization problem. In addition, two gradient neural networks are employed for comparison. Theoretical analysis of convergence is presented to demonstrate the exponential convergence of the proposed new recurrent neural network. Simulation results based on computer further demonstrate the efficacy and advantages of the proposed new recurrent neural network, compared with two gradient-based neural networks.
{"title":"Three neural networks for nonlinear optimization","authors":"Mei Liu, Ran Yang, Bolin Liao","doi":"10.1109/ICNC.2014.6975813","DOIUrl":"https://doi.org/10.1109/ICNC.2014.6975813","url":null,"abstract":"The online solution of optimization (including minimization and maximization) is viewed as a basic and important issue, which has been widely arisen in scientific researches and engineering applications. In this paper, a new recurrent neural network (NRNN) is generalized and investigated for the nonlinear optimization problem. In addition, two gradient neural networks are employed for comparison. Theoretical analysis of convergence is presented to demonstrate the exponential convergence of the proposed new recurrent neural network. Simulation results based on computer further demonstrate the efficacy and advantages of the proposed new recurrent neural network, compared with two gradient-based neural networks.","PeriodicalId":208779,"journal":{"name":"2014 10th International Conference on Natural Computation (ICNC)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114856858","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：